In this fiscal year, we have focussed on defining the functional role of myositis autoantigens in muscle differentiation and repair. As in other systemic autoimmune diseases, patients with myositis often have autoantibodies targeting self-proteins. We previously demonstrated that the dermatomyositis (DM) autoantibody CHD4 is upregulated in regenerating muscle and that reducing levels of this protein in cultured muscle cells accelerates their differentiation. Recently, we showed that another DM autoantigen, TRIM33, is similarly up-regulated during muscle repair and seems to play a role during muscle differentiation in cell culture. During this year, we have sought to extend these findings by generating mice that have no CHD4 or TRIM33 expressed in muscle cells. We have generated the first litters of such mice and will work in the coming fiscal year to analyze how deletion of these genes in muscle cells affects muscle cell development and regeneration following muscle injury. In addition to the basic science project described above, we also have been involved with several clinical projects involving myositis patients who are part of the Johns Hopkins Myositis Center Longitudinal Cohort. These projects include (a) analyzing a longitudinal cohort of patients with anti-HMG-CoA reductase antibodies to determine optimal treatments modalities, (b) demonstrating that unique patterns of muscle biopsy abnormalities are associated with individual dermatomyositis autoantibodies, (c) defining the spectrum of diseases associated with anti-NT5C1a autoantibodies and defining the associated clinical features in each disease, and (d) defining the clinical features of myositis patients with different autoantibodies. In collaboration with Dr. James Gulley and colleagues at the NCI, we have sought to characterize and identify pathogenic mechanisms in patients with metastatic thymoma who develop myositis following treatment with PD-L1 checkpoint inhibitors. In collaboration with Dr. Avi Nath and colleagues at NINDS, we helped characterize a mouse model for motor neuron disease (MND) created by over-expressing the endogenous retrovirus HERV-K. We showed that muscle from these mice had fiber-type grouping as seen in MND.